Thanks for that. I do understand about the dead-head.

Just to summarize where we are, since some fog may be starting to settle over the thread:

We have a working simulation of manual-latch operation. It uses working with the petcock to simulate having the latch hoses disconnected and operating the latch manually. It requires no electrical mods of any type. No resets have been required. Tested over many raise/lower top cycles with no "convertible top not latched" or other error messages. A little work still to do on operation of the windows when raising the top, but that looks manageable.

In a nutshell, you could say this is a spoof of the system ... getting it to think it is controlling the latch ... rather than an attempt to re-cast the system box-by-box for manual latch operation.

What's left is to go from simulation to a working system. That will involve disconnecting the latch hoses and terminating the latch ports on the pump in some way. And also documenting a simple procedures for operating the top with a manual latch.

As time permits, I think I'll just go ahead and convert my own car -- at least temporarily -- if there's nobody with failed hoses currently looking for an alternative to replacing them.

 

The latch hose output ports are 1/8 BSP.
I suggest connecting them together since one will always be Flow and the other Return so the pump will develop no pressure if it isn't disconnected (I believe early models had connections to the solenoids but Gus warns that later ones can very easily be damaged.
You can connect the open ends of the hoses together but it serves no purpose and will give some back pressure since there is a volume difference between the two sides of the ram due to the rod displacement.

 

Steve,

Thanks.

Give a yell please if you disagree with any of the following.

Latch ports on the pump: Yes, connected together somehow ... or maybe capped. I had originally thought just a shunt hose. But I'm worried this would pull down the pump pressure and mess up raising/lowering the top (there will be flow there while the top is moving). Cap the ports instead? Potential dead-head issues this way, but I think it would definitely be OK for a pump with a relief valve installed. Less clear, but still OK I think with a resistor installed. For a stock pump, best option looks to be a shunting hose with an orifice to limit flow so as not to lose too much pump pressure.

Solenoids: I did not intend to mess with them at all. It looks to me to not be required. Change the fewest things possible.

Abandoned latch hoses: Right, not connected together. As you said, If there were still fluid in them, we'd have unequal flow rates because of rod displacement. Let 'em breathe.

My car: a resistor is installed, and it's my neck so-to-speak, so I think I'll cap the pump ports initially and do some testing. Question is ... WHEN? Great; I've just created another project.

 

Dennis,
I wouldn't cap the ports. The pump won't like that.
The pump only delivers to the latch hoses when in latch/unlatch mode so it won't affect your raise/lower pressure at all.
Good luck, the Force be with you !

 

OK, I understand your concern on the caps.

Please suspend disbelief for just a moment on the following. I thought until recently that the solenoids were on/off switches. But I have become (almost) convinced that they are in fact directional switches .... that both latch and ram hydraulic circuits are pressurized whenever the pump is running. The solenoids determine in which direction the pressure is directed, but it is always there in both circuits.

[edit] To be more precise ... It looks to me that:
- the solenoids are not on/off switches, but directional switches
- there is pressure available to both latch and ram circuits whenever the pump is running
- For each circuit, when the direction of the pump rotation matches the direction setting of the solenoid for that circuit, there will be pressure in the circuit (a sort of logical "and" between the pump direction and the solenoid setting).
Some times both circuits are pressurized; sometimes one or the other.


At most times the pressure in one circuit or the other is ineffective, e.g. pushing on a latch already at its limit as the top is going up/down. In plowing through that big 501-something TSB with all the sequence charts on convertible top events, I can't see any other way to interpret things.

OK, throw a net over me if you must. But take a look and see what you think, OK?

 

The pump design is beyond belief.
There are two things going on
When the Top Up Relay is energized, the hydraulic pump is in convertible top raising mode (push direction - pistons extend).
When the Top Down Relay is energized, the hydraulic pump is in convertible top lowering mode (pull direction - pistons retract).
AND
When energized, the Latch Control Valve allows fluid to be pumped into the base of the latch
mechanism cylinder to extend the piston. When de-energized, the Latch Control Valve allows fluid to exit the base of the latch mechanism cylinder to allow the piston to retract.
When energized, the Main Control Valve allows fluid to be pumped into the base of the convertible top cylinders to extend the pistons.
When de-energized, the Main Control Valve allows fluid to exit the base of the convertible top cylinders to allow the pistons to retract.

So you should be right. There's no mechanism to switch between latch and lift rams.
If the TSB is correct then I also don't see a mechanism to open the rod side of either ram to Return.
Looking at the pressure traces in Gus's article
http://www.jagrepair.com/HydPressureReliefValve.htm
there remains pressure in the latch ram while the top rams are moving - so maybe capping those ports is the right thing to do.
1/8 BSP blanking plugs + Dowty washers are cheap enough - or you may find a female 1/8 BSP cap to screw onto the existing adapter..

 

Hey, thanks much for that, I was doubting my own sanity (in this I have company).

For example, "Top Closing Sequence" step 5 (page 6, bulletin 501-11) ... both control valves are open as the top is raising ... both ram and latch circuits are pressurized. In a "normal" car, the latch is already extended so no flow there, no pressure loss. But if we shunt the latch circuit for the manual-latch spoof, then we'll pull down the pressure needed for raising the top.

I have like zero practitioner skills with the components you mentioned. But I'll read up. I think I can start by capping things, if I'm careful while testing.

 

It takes a bit of thinking through - at the start of the 'Open' sequence the pump is running 'forwards' and forcing the top rams hard against their 'fully extended' position and also the latch ram to open.
Once the correct sequence of events from the latch microswitches have been received the pump reverses.
This is the tricky bit to understand - the top's now trying to lower but the latch is also trying to close which would trap the top.
Presumably the pressure required to close the latch is greater than the pressure to get the top moving so it escapes the latch, lowers, hits the stop in the fully down position and the pressure rises enough to close the latch.
The closing sequence is more straightforward.
Interesting.

 

Hey Doug,
This is perhaps the most depressing post I have read since I joined the forum. You avoid using the convertible so it won't break; isn't the reason you bought this vehicle while living in Los Angeles, specifically so you could enjoy about 355 days a year in the sunshine with the top down? You're depriving yourself in anticipation? I also had the green shower, first from a burst hose and then from the actuator valve, and then after putting in stainless steel hi pressure hoses, continued to have leaks to the point that I took it back to the shop 16 times until it finally stopped. I was angry, screaming, frustrated....but I wasn't going to let it stop me from enjoying the reason I bought my first convertible. Don't let the trepidation of what might happen keep you from fulfilling your dreams of topless driving, it would be a shame. I put the top down everyday that it's not raining, cold or hot. I'll never be without a 'vert' in my collection ever again. Enjoy!

 

The ultimate answer to me was buying the Coliflower braided hose kit to replace the latch hoses. I have no fear of using my top and I have no fear of that hose ever failing.

 

No it goes from operating the latch control valve to the lift control valve by means of the switches incorporated in the system located in the header and rams.

 

Well, Gus that's what I thought but I now think Dennis is correct and the system pressurises both lift and latch all the time.
Where did you put the pressure tap for the graph in your article? It clearly shows ~300 psi in between the top latch open and close pressure spikes.
I can't see any mechanism for diverting the pressure from latch to lift.

 

It was tapped into the unit below the two hoses. Any pressure indicated could be residual. The way I read the 501-11 you will see that a switch must change status to move on to the next sequence that is the problem everyone has with this system. I do not think this pump is designed to control both the roof and latch at the same time.

 

I agree Jude the Obscure should have written that TSB.
The only way I can make the pump work with just two solenoid valves is to supply both lift + lock rams simultaneously.
If you look at your trace there's a 'bump' in the middle which is the point of maximum force on the lift rams. If I really reduce the blow-off on the relief valve I can stall the top there.

I think they reverse the pump because that's cheaper than doing it properly - run the pump 'forwards' all the time and use proper electric spool vales for independent control of lock and lift.

You know more about this pump than I do and I'm more than happy to be wrong but at the moment I'm a convert to the School of Dennis.

 

The terminology in our friend TSB 501-11 could hardly be more confusing, the top-closing sequence chart on pg. 6 being a good example.

The Main and Latch control valve states are given as "ON" and "OFF", but ON really means "set to permit flow to Extend the ram(s)", and OFF means "set to permit the opposite flow, to Compress the ram(s)". Pg. 2 documents this. It's helpful to relabel the chart E and C vs. On and Off.

Neither valve is EVER "closed"! I don't think such a state exists.

I'm pretty sure of it all now, and this is repeated (with some scrubbing) from post above:
- the Main and Latch valves (solenoids) are not on/off switches, but directional switches. One direction is to Expand the rams, the other to Compress them.
- there is pressure available to both Main and Latch circuits whenever the pump is running.
- For each circuit, when the direction of the pump rotation (Expand or Compress) matches the direction setting of the valve for that circuit, there will be pressure in that circuit.
Sometimes both circuits are pressurized; sometimes just one or the other.

So for example in the 501-11 chart, pg. 6...
- in step 4, the Latch circuit is pressurized, the Main circuit not,
- in step 5, both circuits are pressurized.

Steve, all of the pressure measurements that I'm aware of have been done through a tap in the pump body. So they show the pressure the pump is creating whether applied to one or the other or both hydraulic circuits.

 

I really don't know.
Gus had a pump to bits to work out where to put the relief valve.
I haven't so I can't say - it would mean tapping the pressure off each port to see what really happens.
I don't think your latest doctrine is the True Word because there's no 'Return' flow with your system.
I'll cogitate over a glass or two of RM !

 

xenophobe,

Your braided hoses may never fail, but there are still a number of weak links in this Rube Goldberg-like chain. Pumps, fittings, microswitches, and everything in between all add up to multiple points of potentially expensive failures. I still believe that a simple manual top with interior handle-type latches on the driver side and the passenger side would be the ideal solution....

 

Steve,

Sorry, I don't know what you mean by "no return flow"? Can you 'splain a little more?

edit: Forgot ... pg, 2, second paragraph, is unambiguous. Both the ON and OFF states of the valves "permit flow". It's the permitted direction that changes.

 

I can try.
On a normal hydraulic system you've got a series of spool valves which have three positions 1-R-2
At rest (in 'neutral') all the valves are in the 'R' position so the pump output goes straight back into the reservoir, i.e; the pump does no work.
As soon as any valve is moved to (e,g,) '1' the return flow is cut off and the pump output is diverted to side '1' of the ram. At the same time side '2' of that ram is connected to the reservoir so fluid pushed out of side '2' by the piston goes back into the reservoir.
To do this with solenoid valves requires either a three port valve with some fancy plumbing or two 2 port valves per cylinder.

 

I did start using the convertible more finally as the weather got better but then we had that cold snap (for SoCal) and then the rains. Now that the weather is getting nicer again I will try and push myself to get that top down.

By the way, what shop were you using that it took 16 times to get the theoretically leak-proof hoses properly installed?

Doug